This is a study of the molecular mechanism of genetic translation, which is an essential process for the health and well-being of all organisms. Recent work shows that codon specificity is partly determined by tRNA structure. The proposed work will elucidate how tRNA an other translational components determine decoding specificity. Variants of E. coli tRNA2Arg, which uses inosine to decode CGA, CGU and CGC, will be studied. Preliminary analyses show that decoding with an adenine:inosine (A:I) base pair is inefficient. Clusters of nine CGA (but not CGU) dramatically reduce expression. Detailed in vivo analyses show that (1) the requirement for an A:I base pair slows aminoacyl-tRNA selection, an (2) when in the P site A:I interferes with translation of the next triplet. These decoding defects provide a rationale for the low usage of codons translated with A:I in virtually all organisms. The principal experimental approach will be to clone, as oligonucleotides, variants of the tRNA genes will specify either I, U or G as the 5' anticodon nucleotide. Each synthetic gene will be mutagenized by partially randomized synthesis at positions outside the anticodon. The host chromosome will be mutagenized in another study. Mutagenized tRNA genes or hosts will be selected for efficient expression of clustered CGA codons. Translational efficiencies of mutants will be compared at CGA, CGU and CGC. The results are expected to illuminate features of the decoding apparatus that determine decoding specificity and efficiency. The tRNA variants with anticodon U, which will decode with an A:U pair, may also explain why no natural rRNA decodes CGA with U. Variants that translate with an A:G pair will define structural constraints that prevent this normally errant A:G decoding. These experiments will be performed on sense codons in vivo. Therefore, all results must be relevant to the cellular translational repertoire. It is also expected that because the decoding mechanism is essentially the same in all organisms, the results will be generally applicable for understanding this fundamental molecular process. The methods for making E. coli variants that have novel and enhancing decoding features will be useful for expression of medically important genes in microorganisms. Finally, this work will attract and train students for careers in biomedical research.